To persist or cache results between flow executions in Prefect, you don't directly use the Prefect API, as it does not store your results except in specific cases. Instead, the result is saved to a storage location in your infrastructure, and Prefect keeps a reference to it. This functionality is important for task cache keys and flow run retries.
Configuring persistence requires a serializer and a storage location. Prefect provides defaults, which you typically wouldn't need to change unless you want to customize behavior. You can configure results on the task or flow level.
For detailed steps and configuration, please refer to the
Prefect documentation on persisting results.